Vitis 34 (2), 119-123 (1995)

Transmission of grapevine is not likely to occur mechanically by normal pruning

by

1 U. STAUB, H. PouvKA, J.V. HERRMANN ) and H.J. GRoss

Bayerische Julius-Maximilians-Universitat, Institut fiir Biochemie, Biozentrum,Wiirzburg, Germany 1 ) Bayerische Landesanstalt fiir Weinbau und Gartenbau, Veitshochheim, Germany

S u m m a r y : In epidemiological studies the distribution in two local vineyards was determined. Grapevine leaves of different varieties were collected, total RNA isolated and viroid detected by northern blot analysis and/or reverse transcription followed by PCR amplification. Nearly each sample was infected with the grapevine variant of and approximately each second additionally with Grapevine Yellow Speckle Viroid l. Grapevine Yellow Speckle Viroid 2, a third grapevine viroid, was not found. Both grapevine viroids occurred in chlorotic plants as well as in plants without symptoms. In order to investigate viroid spreading through mechanical transmission accomplished during routine cultural practices, the distribution patterns in the two vineyards were analysed. Our results indicate that grapevine viroids are mainly propagated through systemic transmission upon grafting. The examination of different rootstock clones from Northern Italy, which are used for grafting in Germany, further demon­ strates that in this case propagation is not due to rootstocks containing viroid but is more likely to occur via infected scion varieties.

K e y w o r d s : grapevine viroids, northern blot analysis, RT/PCR, rootstock, vine variety.

Introduction has been isolated from Vitis vinifera cv. Riesling (PucHTA et al. 1988) and the rootstock 5BB (PucHTA et al. 1989). Viroids are the smallest known pathogens infecting only We performed this study with the aim to determine higher plants (RIESNER and GRoss 1985). They exist as cir­ the viroid distribution in two local vineyards. The analy­ cular, single-stranded RNA molecules with a length of 240- sis of viroid distribution patterns within a vineyard and in 600 nucleotides. Viroids of grapevine are distributed world­ different rootstock clones from Caprino Veronese (North­ wide and have been detected and characterized in Australia em Italy), which are routinely used for grafting in Ger­ (KoLTUNow andREzAIAN 1988 and 1989; REZAIAN et al. 1988; many, allows an insight into the mechanisms of viroid REZAIAN 1990), Germany (PUCHTA et al. 1988 and 1989), spreading. Japan (SANO et al. 1985), California (SzYcHowsKI et al. 1988) and Spain (GARCiA-ARENAL et al. 1987). From the five Materials and methods known grapevine viroids HSVdg (Hop Stunt Viroid grape­ vine), GYSVd1 (Grapevine Yellow Speckle Viroid 1), P 1 an t s : In a local vineyard (vineyard A) young GYSVd2 (Grapevine Yellow Speckle Viroid 2) that was leaves from Vitis vinifera cvs Bacchus, M tiller-Thurgau formerly named GV1b, CEVdg (Citrus Exocortis Viroid (rootstock 5 C), Kemer and Silvaner (rootstock SO 4) were grapevine) and AGVd (Australian Grapevine Viroid), only collected and stored at -20 oc until use. During the hot GYSVd1 and GYSVd2 were reported to induce symptoms summer of 1993 plants of cvs Bacchus and Miiller-Thurgau after infection ofviroid-free grapevine seedlings (KoLTUNOW were severely chlorotic whereas plants of cvs Kemer and et al. 1989). Grapevine viroids belong to the two main viroid Silvaner expressed no symptoms. For comparison we col­ groups that can clearly be distinguished from each other lected leaves of cvs M tiller-Thurgau and Kerner on by differences in their central conserved region (CCR) rootstock SO 4 in another local vineyard (vineyard B). (KEESE and SYMONS 1985). HSVdg is a member of the Even in the hot summer 1993 these plants expressed no PSTVd (Potato Spindle Tuber Viroid)-group and GYSVd1, chlorotic symptoms. In order to elucidate whether viroid GYSVd2 and AGVd are members of the ASSVd (Apple infection and spreading derives from the rootstock, leaves Scar Skin Viroid)-group (HASHIMOTO and KoGANEZAWA from different clones in Caprino Veronese (Northern Italy) 1987). In comparative studies between Europe and Cali­ were also obtained and assayed for viroid infection. fornia (SzYCHOWSKI et al. 1991) and California and Aus­ tralia (REZAIAN et al. 1992) it turned out that nearly all sam­ E x t r a c t i o n o f n u c 1 e i c a c i d s : Grapevine ples studied were infected with the latent and symptomless leaves (0.25 g samples) were ground in liquid nitrogen HSVdg (SANO et al. 1985). Furthermore, many cultivars with a mortar and a pestle. The powder was transferred were coinfected with either GYSVd1 or GYSVd2, or with into a 2.0 ml microcentrifuge tube and 800 111 homogeni­ both. Plants solely infected with GYSVd1 or GYSVd2 with­ zation buffer (0.2 M boric acid, 10 mM Na2EDTA, pH out a HSVdg-infection were not found. In Germany HSVdg adjusted to 7.6 with Tris), 16 111 25% SOS and 16 111 of

Correspondence to: Prof. Dr. H. J. GRoss, Universitat Wiirzburg, Institut fiir Biochemie, Biozentrum, Am Hub land, D-97074 Wiirzburg, Germany 120 U. STAUB, H. PoLIVKA, J.V. HERRMANN and H.J. GRoss

2-mercaptoethanol were added. The mixture was shaken nucleotide primers, were employed for RT/PCR (STAUB et for 10 min with 800 ~1 phenol/chloroform/isoamyl alco­ al. 1995). Amplification products were separated in a 2.0 % hol (25/24/1, v/v/v) saturated with 10 mM Tris-HCl/1 mM agarose gel and visualized by ethidium bromide staining.

Na2EDTA (pH 7.0) and centrifuged at 15000 g (4 °C) for 10 min. Phenol extraction was repeated and the aqueous N o r t h e r n b I o t a n a 1 y s i s f o r t h e d e­ phase finally extracted with 400 ~1 chloroforrn/isoamyl t e c t i o n o f g r a p e v i n e v i r o i d s : 5 jlg total alcohol (24/1, v/v). Nucleic acids were further purified ei­ RNA was separated in a 5% polyacrylamide/7 M urea gel ther by differential solvent precipitation (MANNING 1991) and electrotransferred onto a positively charged nylon or by an aqueous two-phase system (BEUTIIER et al. 1988) membrane (Qiabrane Nylon Plus; Qiagen, Hilden). Hybridi­ followed by DEAE-cellulose chromatography according zation was carried out at 60 °C with equal amounts of ra­ to STAUB et al. (1995). dioactively 5' end-labelled oligodeoxynucleotides HV1 and GV3 simultaneously (Tab. 1; STAUB et al. 1995). After Detection of viroids by RT/PCR: In hybridiziation the membrane was washed under stringent a coupled reaction (MYERS and GELFAND 1991) viroid-RNA conditions (0.1 x SSC, 0.1% SDS; 60 °C) and the result in 25 ng total RNA, isolated from grapevine leaves, was was documented by autoradiography. Signal intensities reverse transcribed (RT) and viroid-cDNAamplified (PCR). were quantified with a Molecular Dynamics Phosphor­ GV1 and GV2 (Tab. 1), two GYSVdl-specific oligodeoxy- Imager 425.

Tab I e I Oligodeoxynucleotide primers used for RT/PCR studies and northern blot analyses

name sequence specific for position within viroid domain the molecule

GVl 5'-GCGGGGGTTC CGGGGATTGC-3' GYSVdl 341 - 360 TL

GV2 5'-taagaggtct ccggatcttc ttgc-3' GYSVdl 361- 17 TL

GV3 5'-ACCCCTICGT CGACGACG-3' GYSVdl 98- 115 CCR GYSVd2

HVl 5'-GTIGCCCCGG GGCTCCTI-3' HSVdg 71 - 88 CCR

Sequences in capital letters are complementary to the corresponding viroid, the sequence of GV2 (in lower case letters) is the DNA homologue of the viroid sequence. GYSVdl: Grapevine Yellow Speckle Viroid 1; GYSVd2: Grapevine Yellow Speckle Viroid 2; HSV dg: Hop Stunt Viroid grapevine; T L: left terminal region of the molecule and CCR: central conserved region of the viroid according to KEESE and SYMONS (1985).

Results Samples of other varieties in vineyard A were exam­ ined in the same way. The results are summarized in Fig. 3. Viroid distribution in two local Nearly each plant is infected with HSVdg and about 50% v i n e y a r d s : In the investigated areas two of the five of the samples studied are additionally infected with known grapevine viroids, HSVdg and GYSVd1, were iden­ GYSVdl. 81% of chlorotic cv. Bacchus (rootstock 5 C) tified by RT/PCR and northern blot analyses. For the de­ samples and 46% of chlorotic cv. Muller-Thurgau tection of GYSVd1 in different samples RT/PCR was car­ (rootstock 5 C) samples, but only 29% of symptomless cv. ried out with the specific oligodeoxynucleotides GV1 and GV2. The expected PCR products with 367 bp length cor­ M 23456 7 responding to full-lenght GYSV dl were visualized by stain­ 4~9--.,.,._ ing with ethidium bromide (Fig. 1, lanes 3-7). 41~----­ In this epidemiological study grapevine viroids were .1 .'1- - 367bp 2-l~_../ mainly detected by northern blot analyses as shown in 1'){)__...... 147 ___...... Fig. 2. 5 ~g total RNA isolated from different samples of I 10 ___...... cv. Muller-Thurgau in vineyard A were fractionated by electrophoresis in a 5% polyacrylamide/7 M urea gel, Fig. I: Detection of GYSVdl by RT/PCR from young electrotransferred onto a positively charged nylon mem­ leaves of different grapevines. Lane M: marker pUC19/Hpa II fragments brane and hybridized with labelled oligodeoxynucleotides (length in bp). Lane 1: control; RT/PCR product from total RNA GV3 and HVl. Some of the assayed samples are viroid­ of a GYSVdl-infected plant (cv. Mtiller-Thurgau). Lane 2: con­ free (Fig. 2, lanes 4, 12) and others are infected with either trol; RT/PCR with ddHp instead of total RNA. Lanes 3-7: RT/ HSVdg (lanes 6, 9, 11, 13) or HSVdg and GYSVdl (lanes PCR products from total RNA of different samples of 3, 5, 7, 8, 10). cv. Traminer. Transmission of grapevine viroids 121

Kemer (rootstock SO 4) and 37% of symptomless Silvaner The investigation in vineyard B (symptomless plants (rootstock SO 4) samples are positive for both viroids. of cvs. Kerner and MUller-Thurgau on rootstock SO 4) GYSVdl and HSVdg were isolated from both symptomless showed that 65% of the cv. Kemer samples were infected and chlorotic plants, demonstrating that the observed with GYSVdl but only 25% of the neighbouring cv. MUller­ chlorosis is not simply related with viroid infections. Thurgau samples (Fig. 5). Moreover, only 46% of the GYSVdl-infected plants are always coinfected with HSVdg cv. MUller-Thurgau samples were infected with HSV dg. but not vice versa (Fig. 4). Samples solely infected with For comparison the infection rate with HSVdg in the GYSVdl were not detected. cv. Kerner area of vineyard B was 100% and approximate! y 95% in vineyard A (Fig. 3). 2 3 4 s 6 7 8 9 10 11 12 13 oYSVdl,.­ HSVcl&..:­ OYSVdJ.I- HSVcta-1-

Fig. 2: Northern blot analysis for the simultaneous detection of grapevine viroids HSVdg and GYSVdl. Lane l: control, total RNA from a double-infected grapevine cultivar. Lane 2: control, RNA from a HSVdg and GYSVdl infected grapevine cultivar. Lanes 3-13: total RNA isolated from grapevine leaves of cv. variety MUller-Tburgau Kemer grafted on rootstock S04 504 MUller-Thurgau. c: number of assayed samples 24 20 circular form of the molecules; 1: linear form chlorosis of HSVdg and GYSVdl. Fig. 5: Relative occurrence of HSVdg and GYSVdl in vineyard B. Plants of cvs. MUller-Thurgau and Kerner (rootstock SO 4) in this vineyard were symptomless. Detection of viroids was car­ ried out as mentioned in Fig. 3. Note the unexpected large portion of completely viroid-free cv. MUller-Thurgau plants.

I n f e c t i o n p a t t e r n s a n d v i r o i d p r o­ p a g a t i o n : Analyses of viroid distribution patterns in different areas of vineyard A, results shown for samples of cv. MUller-Thurgau (Tab. 2), indicate that grapevine viroids are not mechanically transmitted by normal pruning for the following reasons: HSVdg was detected in nearly each variety Bacchus MUller-Thurgau Kemer Silvlner grafted on rootstock se ~c S04 S04 sample confirming the expectation of a worldwide distrib­ number of assayed samples 48 24 24 24 chlorosis ++ uted latent grapevine viroid (REZAIAN et al. 1992; Fig. 3: Relative occurrence of HSVdg and GYSVdl in different SZYCHOWSKI et al. 1991). GYSVdl, however, is randomly areas of vineyard A The different stages of chlorosis are indi­ distributed in both vineyards and only each second sample cated by ++: severe chlorotic, +: chlorotic, -: without chlorosis. was positive. If viroid transmission would occur mechani-

Table 2

Pattern of distribution of grapevine viroids in samples of Vitis vinifera cv. MUller-Thurgau from vineyard A

row/plant HSVdg-infected variety B acchus MUller· Thuraau GYSVdl-infected grafted on rootstock !5C lC number of assayed samples 48 24 chlorosb ++ 112 + + 116 l/10 + + l/l4 +

4/2 + + 4/6 + + 4/10 + variety Kemcr Silvaner grafted on root1tock S04 S04 4/l4 + + number of assayed samples 24 24 chlorosis 16/2 + + Fig. 4: Patterns of distribution of grapevine viroids in samples 16/6 + of vineyard A. Dark areas indicate the portion of plants infected 16/IO + with HSVdg and GYSVdl, white areas the portion of viroid-free 16/14 + plants. Plants infected solely with GYSVdl were not found. 122 U. STAUB, H. PouvKA, J.V. HERRMANN and H.J. GRoss cally by pruning, neighbours of GYSVdl-infected plants reproduced for sale to Germany, leaves from clones of dif­ should also be infected. Moreover, there are a number of ferent rootstocks (5 C Geisenheim, Kober 125 AA completely viroid-free plants in the area of cv. Muller_. Geisenheim, SO 4, Kober 5 BB, Binova) were collected Thurgau of vineyard B (Fig. 5). Therefore mechanical trans­ and assayed for viroid infection (Tab. 3). Northern blot mission during routine cultural practices is rather unlikely. analyses revealed that only 8 of 56 samples were HSVdg­ Distribution of grapevine viroids infected. Neither GYSVdl nor GYSVd2 were detectable. in clones of different rootstocks: In This result indicates that spreading of grapevine viroids is Caprino Veronese (Northern Italy), where rootstocks are not likely to occur via viroid-containing rootstocks.

Table 3 Occurrence of viroids in grapevine leaves of different rootstock clones from Caprino Veronese (Northern Italy)

rootstock clone no. number of HSVdg- GYSVdl- GYSVd2- assayed samples infected infected infected

5CGm 6-2-13 5 2 0 0 6-4-22 2 1 0 0 6-5-52 1 1 0 0 6-7-16 3 3 0 0 27-7 4 0 0 0

Kober 125 AA Gm 3 5 1 0 0 5 3 0 0 0

S04 31 Opp. 11 0 0 0

Kober 5BB 13-46-15 2 0 0 0 13-46-13 5 0 0 0 13-45-5 4 0 0 0 13-44-21 5 0 0 0

Binova 25020117 5 0 0 0 25020217 1 0 0 0

infected 8 of 56 0 of 56 0 of 56

Gm: Geisenheim

Discussion was quantitatively determined with a Phosphorlmager af­ ter hybridization with specific probes, was normally four­ Although only a limited number of samples compared fold higher than that of HSVdg (STAUB et al. 1995), corre­ with the total number of plants in the two vineyards was sponding to the results of earlier investigations (SzYCHOWSKI examined, random spot checks in both vineyards confirmed et al. 1988; MINAFRA et al. 1990; SZYCHOWSKI et al. 1991). our notion that viroid transmission does not occur via the In some chlorotic samples of cv. Muller-Thurgau (vine­ regular pruning. The detection of grapevine viroids was yard A), however, the GYSVd1/HSVdg ratio was 10:1 to possible throughout the vegetation period. Nearly each 20:1 (Fig. 2, lanes 5, 7, 10). Further investigations have to sample was HSVdg-infected and approximately each sec­ prove whether this accumulation of GYSVd1 depends on ond additionally with GYSVd1 (Fig. 3). Grapevine viroids certain GYSVd1- or HSVdg-variants in special varieties were isolated from both symptomless and chlorotic plants, or whether it correlates with symptom expression. demonstrating that there is no simple correlation between Two main mechanisms have been discussed for viroid viroid infection and symptom expression as mentioned spreading (SzYCHOWSKI et al. 1988): (i) mechanical trans­ earlier (SANO et al. 1985; KOLTUNOW and REZAIAN 1988). mission accomplished during routine cultural practices and We found no plants solely infected with GYSVdl (Fig. 4), (ii) systemic transmissions by either infected rootstocks or confirming the results of other studies carried out in Italy scion varieties. According to our study the first possibility and California (MrNAFRA et al. 1990; SzYCHOWSKI et al. does not appear to be likely. As expected, HSVdg is present 1991 ). Contrary to these studies we never detected in nearly each plant. Far more interesting is the distribu­ GYSVd2. The GYSVdl-titer in grapevine leaves, which tion of GYSVd1 that appears in each second sample. Transmission of grapevine viroids 123

GYSV d1-infected plants do not form clusters within a vine­ KEEsE, P.; SvMoNs, R. H.; 1985: Domains in viroids: evidence of intermo­ yard as it should be expected if this viroid is transmitted lecular RNA rearrangements and their contribution to viroid evolu­ tion. Proc. Natl. Acad. Sci. USA 82, mechanically by the regular pruning 4582-4586. (Tab. 2). KoLTUNOW, A. M.; REzAIAN, M. A.; 1988: Grapevine yellow speckle viroid: In contrast to earlier suggestions (PucHTA et al. 1989) structural features of a new viroid group. Nucleic Acids Res. 16, 849- we could show that transmission through viroid-contain­ 864. ing rootstocks plays a minor role. Only a few of the clones - -;- -; 1989: Grapevine viroid IB, a new member of the apple scar skin viroid group contains the left terminal region of tomato planta ma­ of different rootstocks collected in Caprino Veronese, from cho viroid. Virology 170, 575-578. where local vine growers are supplied with rootstocks, were - -; KRAKE, L. R.;JOHNSON, S. D.; REzAIAN, M. A.; 1989: Two related viroids infected with HSVdg (Tab. 3). Moreover, neither GYSVd1 cause grapevine yellow speckle disease independently. J. Gen. Virol. nor GYSV d2, the two grapevine viroids that can cause 70, 3411-3419. MANNING, K.; 1991: symptoms (KoLTUNOW Isolation of nucleic acids from plants by differential et al. 1989), were detected in these solvent precipitation. Anal. Biochem. 195, 45-50. clones. Although the number of samples was limited there MINAFRA, A.; MARTELU, G. P.; SAVING, V.; 1990: Viroids of grapevine in is no evidence that local grapevine viroids originate from Italy. Vitis 29, 173-182. infected rootstocks. Systemic transmission by infected MYERS, T. W.; GELFAND, D. H.; 1991: Reverse Transcription and DNA ampflification scion varieties by a Thermus thermophilus DNA polymerase. Bio­ is therefore most likely and could be the chemistry 30, 7661-7666. best explanation for our results obtained with samples from PucHTA, H.; RAMM, K.; SANGER, H. L.; 1988: Nucleotide sequence of a hop vineyards A and B. The varieties Kerner and Miiller­ stunt viroid isolate from the German grapevine cultivar 'Riesling'. Thurgau of vineyard B are both grafted on rootstock SO 4, Nucleic Acids Res. 16, 2730. - -;- -; LUCKINGER, R.; FREIMOLLER, K.; SANGER, H. L.; 1989: but only in the Kemer area grapevine Nucleotide viroids are spread to sequence of a hop stunt viroid (HSV d) isolate from the German grape­ a great extent. Moreover, in all investigated areas single vine rootstock 5BB as determined by PCR-mediated sequence analy­ uninfected plants exist among infected neighbours (Tab. 2). sis. Nucleic Acids Res. 17, 5841. This should not be the case if viroid transmission occurs REZAIAN, M. A.; 1990: Australian grapevine viroid- evidence for exten­ sive recombination between viroids. Nucleic Acids Res. 18, 1813- by the regular pruning. Hence, viroid transmission appears 1818. to be caused by the use of viroid-infected scion varieties - -; KoLTUNOW, A. M.; KRAKE, L. R.; 1988: Isolation of three viroids and a for grafting. circular RNA from grapevines. J. Gen. Virol. 69, 413-422. - -; KRAKE, L. R.; GouNo, D. A.; 1992: Common identity of grapevine viroids from USA and Australia revealed by PCR analysis. Acknowledgements Intervirology 34, 38-43. RIESNER, D.; GRoss, H. J.; 1985: Viroids. Ann. Rev. Biochem. 54, 531-564. This work was supported by Bayerisches Staatsministerium SANO, T.; 0HSHIMA, K. ; UYEDA, I.; SHIKATA, E.; MESH!, T.; 0KADA, Y.; 1985: ftir Ernahrung, Landwirtschaft und Forsten and by Fonds der Nucleotide sequence of grapevine viroid: a grapevine isolate of hop Chemischen Industrie. stunt viroid. Proc. Japan Acad. 61, Ser. B, 265-268. STAUB, U.; PouvKA, H.; GRoss, H. J.; 1995: Two rapid microscale proce­ dures for the isolation of total RNA from leaves rich in polyphenols and polysaccharides: application for the sensitive detection of grape­ vine viroids. J. Virol. Method. (In press.) Literature cited SzvcHowsKI, J.A.; GoHEEN,A. C.; SEMANCIK, J. S.; 1988: Mechanical trans­ mission and rootstock reservoirs as factors in the widespread distri­ BEUTHER, E.; K6sTER, S.; Loss, P.; ScHUMACHER, J.; RlEsNER, D.; 1988: Small bution ofviroids in grapevines. Amer. J. Enol. Viticult. 39,213-216. RNAs originating from symptomless and damaged spruces (Picea - -; DoAZAN, J. P.; LECLAIR, P.; GARNIER, M.; CREDI, R.; MINAFRA, A.; DuRAN­ spp.). J. Phytopathol. 121, 289-302. VILA, N.; WoLPERT, J.A.; SEMANCIK, J. S.; 1991: Relationship and pat­ GARCIA-ARENAL, F.; PALLAs, V.; FLoRES, R. ; 1987: The sequence of a viroid terns of distribution among grapevine viroids from California and from grapevine closely related to severe isolates of citrus exocortis Europe. Vitis 30, 25-36. viroid. Nucleic Acids Res. 15, 4203-4210. HASHIMOTO, J.; KOGANEZAWA, H.; 1987: Nucleotide sequence and second­ ary structure of apple scar skin viroid. Nucleic Acids Res. 15, 7045- Received November 11, 1994 7052.